The Effects of Moisture on the Material Properties and ...

Mark R. VanLandingham, J Rushad E Eduljee, 2 and John W. Gillespie, Jr. 3

The Effects of Moisture on the Material Properties and Behavior of Thermoplastic Polyimide Composites REFERENCE: VanLandingham, M. R., Eduljee, R. E, and Gillespie, J. W., Jr., "The Effects of Moisture on the Material Properties and Behavior of Thermoplastic Polyimide Composites," High Temperature and Environmental Effects on Polymeric Composites: 2nd Volume, ASTM STP 1302, Thomas S. Gates and Abdul-Hamid Zureick, Eds., American Society for Testing and Materials, 1997, pp. 50-63. ABSTRACT: Thermoplastic polyimides are a relatively new class of polymers that exhibit high-temperature stability and are useful in composite applications. One such material is Avimid| K3B reinforced with Magnamite | IM7 graphite fibers. This composite system exhibits excellent strength and toughness, retains its strength and toughness after prolonged exposure to elevated temperatures, and resists microcracking at extremely low (liquid nitrogen) temperatures. Further characterization of IM7/K3B thermoplastic composites is focused on evaluating hygrothermal effects on the material properties and behavior. The environmental conditioning test matrix includes three temperatures (20, 40, and 80~ and four relative humidity levels (75, 85, 97, and 100%). Observations and conclusions from these studies include the following: (I) the moisture diffusivity of IM7/K3B has a classic Arrhenius dependence on temperature; (2) the moisture saturation level depends on the relative humidity level to the power of 1.34 with a maximum value of 0.55% by weight; (3) the glass transition temperature T~ is lowered with moisture absorption but is recovered when the sample is redried; (4) the intralaminar fracture toughness G,,. remains constant after extensive hygrothermal conditioning; (5) the diffusion kinetics are Fickian in general, except for a few non-Fickian anomalies that are related to development of transverse microcracks. KEY WORDS: polymer matrix composites, polyimide matrix, graphite fibers, moisture, hygrothermal effects, glass transition temperature, fracture toughness, diffusion, transverse microcracking

Graduate research assistant, Materials Science Program, Center for Composite Materials, University of Delaware, Newark, DE 19716. -"Assistant professor of Materials Science and research associate of the Center for Composite Materials, University of Delaware, Newark, DE 19716. Associate professor of Materials Science and associate director of the Center for Composite Materials, University of Delaware, Newark, DE 19716.

Copyright by ASTM Int'l (all rights reserved); Mon Feb 22 15:54:11 EST 2016 Downloaded/printed by 50 Delaware Univ (Delaware Univ) pursuant to License Agreement. No further reproductions authorized. 9

Copyright 1997 by ASTM International

www.astm.org

VANLANDINGHAM ET AL. ON THERMOPLASTIC POLYIMIDE COMPOSITES

51

Nomenclature

D Transverse diffusivity of the composite laminate (cm2/s) M~,, Moisture saturation level, weight % Gh. Intralaminar fracture toughness, J/m 2 a Notch length of the fracture toughness test specimen, mm 5 Notch width of the fracture toughness test specimen, mm Glass transition temperature, ~ E" Loss modulus, MPa % RH Percent relative humidity of the conditioning environment wt% Weight % or % by weight t~f Fiber volume fraction Over the past 20 years, the effects of moisture on fiber-reinforced polymeric composite materials have been studied extensively [1-9]. In many of these studies, significant changes in the mechanical properties and various forms of moisture-induced damage have been reported [4-8]. For example, absorbed moisture has been shown to reduce the glass transition temperature T~ of the resin [4,5] reduce matrix-dominated properties of the composite such as transverse tensile strength and intralaminar shear strength [4-6], and cause swelling of the resin that induces residual stresses and causes the formation of microcracks [5, 7-10]. These deleterious effects of absorbed moisture have been attributed to plasticization and degradation of the resin matrix and degradation of the fiber-matrix interface [5-10]. To date, most moisture studies have been concerned with thermosetting matrix composites (for example, graphite/epoxy) that absorb as much as 1.2 to 2% moisture by weight in 95 to 100% relative humidity environments (for fiber volume fractions vr between 60 and 68%) [1,2,5-7]. Recently, thermoplastic (semicrystalline and amorphous) matrix composites have been developed that absorb very little moisture compared to their thermosetting matrix counterparts [3,4]. An example of such a system is a thermoplastic matrix composite composed of an amorphous polyimide matrix, Avimid | K3B, reinforced with Magnamite | IM7 graphite fibers (Vf = 62%) [11], that absorbs up to 0.55% moisture by weight during water immersion. In this study, extensive hygrothermal characterization of the IM7/K3B system is conducted to determine moisture-induced degradation, predict moisture contents in service environments, and develop accelerated environmental conditioning methods. Experimental

Materials The particular thermoplastic matrix composite used in this study employed an amorphous polyimide matrix, Avimid| K3B, reinforced with Magnamite | IM7 graphite fibers (Vf = 62%) [11]. Laminates were laid up in house and then processed by The DuPont Co. under optimized processing conditions [12]. Laminate configurations included the following: [90] m, or unidirectional laminates; [02/90]~, or thin cross-ply laminates; [02/902/02]~, or thick crossply laminates; and [45/0/-45/90]~, or quasi-isotropic laminates. After processing and prior to any testing or conditioning, ultrasonic c-scanning (transducer frequency = 15 MHz) was

4 The DuPont Co., Wilmington, DE. 5 Hercules, Inc., Wilmington, DE. Copyright by ASTM Int'l (all rights reserved); Mon Feb 22 15:54:11 EST 2016 Downloaded/printed by Delaware Univ (Delaware Univ) pursuant to License Agreement. No further reproductions authorized.

52

POLYMERICCOMPOSITES: SECOND VOLUME

used to inspect laminate quality. No defects were detected in any of the panels. Also, baseline (dry) longitudinal and transverse tensile testing of unidirectional specimens was performed in accordance with Test Method for Tensile Properties of Fiber-Resin Composites (ASTM D 3039). The average strengths and moduli measured were within 5% of reported values

[I1,13]. Hygrothermal Conditioning The hygrothermal conditioning test matrix consisted of three temperatures, 20, 40, and 80~ and four relative humidity (RH) levels--75, 85, 97, and 100% RH. For 100% RH conditioning, samples were immersed directly in water contained in stainless steel tanks that were roughly 0.003 m 3 in volume. The remaining humidity conditions for 20 and 40~ environments were created within desiccators in accordance with Practices for Maintaining Constant Relative Humidity by Means of Aqueous Solutions (ASTM E 104). For the 40~ conditioning environments, the desiccators were then placed in the water baths used for the immersion studies, while the 20~ desiccators sat out in a controlled room-temperature environment. An environmental chamber was used to create 80~ humidity environments (except for 100% RH). The temperatures of the 40 and 80~ water baths were maintained by a temperature feedback controller, that employed J-type thermocouples (wrapped in a polyimide tape to limit galvanic corrosion) and flexible heating blankets as heating elements. Distilled, deionized water was used in all of the aqueous solutions and in the 100% RH baths. To date, conditioning times of the various specimens range from 1000 to 4400 h. (See the following subsections and the Results section for more specific information on conditioning times.) During preliminary studies of moisture diffusion in the IM7/K3B system, the weights of specimens taken from each type of laminate reached equilibrium (dry) values after 2 to 3 days under vacuum at 120~ Also, in an unpublished study, 6 IM7/K3B specimens were conditioned in 95% RH environments and then dried under vacuum at several elevated temperatures (> 100~ No significant degradation was observed after drying at 120~ Therefore, all specimens were initially dried for 4 days under vacuum at 120~ prior to hygrothermal conditioning. Limited hygrothermal cycling studies were also performed which included 1200 to 1800 h of conditioning in the environments described previously, followed by desorption for 4 days under vacuum at 120~ and finally reimmersion of the specimens into their original environments.

Weight Measurements Square specimens with length dimensions of about 5 cm were cut with a diamond saw from each of the four types of laminates and used to measure moisture uptake during hygrothermal conditioning. At least three samples were conditioned in each constanttemperature, constant-humidity environment for 1200 to 4400 h. A few samples were used for limited hygrothermal cycling studies described in the previous subsection. Prior to conditioning, cross-ply and quasi-isotropic samples were polished on one edge to a 5-p~m finish. This extra step allowed for the nondestructive detection of microcrack initiation by simply observing the polished edge of a sample under an optical microscope each time it was removed for weighing. Weight measurements were taken periodically during both absorption and desorption using a digital balance accurate to +- 1 mg. The surfaces of samples immersed

6 Personal communication with Dick Cornelia, The DuPont Co., Wilmington, DE, 1994. Copyright by ASTM Int'l (all rights reserved); Mon Feb 22 15:54:11 EST 2016 Downloaded/printed by Delaware Univ (Delaware Univ) pursuant to License Agreement. No further reproductions authorized.

VANLANDINGHAM E l AL. ON THERMOPLASTIC POLYIMIDE COMPOSITES

53

in water baths were blotted dry with a cloth before they were weighed. Specimens were removed from their environments for weighing and microcrack inspection for no longer than 30 min (typically 15 to 20 min). Moisture loss during these removal periods was calculated to be insignificant.

Mode I Intralaminar Fracture Toughness Intralaminar fracture toughness Gic was measured using a double edge notch (DEN) test

[14]. Unidirectional panels with dimensions of 20.3 by 13 cm were end-tabbed and cut into four 20.3 by 2.5 cm samples using a slot grinder. Samples taken from conditioned panels were cut at least 1 cm from the panel's edges to eliminate any edge effects from the moisture absorption process. Notches were cut with a 0.5-ram diamond blade to approximately 6.9 mm in length on each side at the center of the sample. Notch lengths a and notch widths were measured to an accuracy of + 0.01 mm using a microscope equipped with a micrometer-driven stage. In Ref 14, notches were cut with a thinner 0.3-mm diamond blade. However, the thinner blade is less rigid and, due to out-of-plane vibration, yields equivalent notch widths to the thicker blade (8 - 0.7 mm). Sample geometry is illustrated in Fig. 1. Baseline (dry) panels were cut and notched and then dried for four days under vacuum at 120~ prior to testing. Samples taken from conditioned panels were cut and notched 2 to 3 days prior to testing. These samples were then placed back into their original environments until they were tested. Total exposure times ranged from 1200 to 3600 h (see Results section), and one panel (four specimens) was conditioned in each environment for each chosen exposure time. All specimens were pulled in tension until failure occurred using a mechanical load frame. All testing was performed at room temperature, and Glc values were calculated as specified in Ref 14. After testing, the fracture surfaces of the G~c samples were observed under a scanning electron microscope (SEM) at magnifications up to 5000X at 30 kV.

2b

a

-I (a)

(b)

FIG. 1--(a) Double edge notch (DEN) test geometry as described in Ref 14 and (b) illustration of actual notch geometry observed in this study. Copyright by ASTM Int'l (all rights reserved); Mon Feb 22 15:54:11 EST 2016 Downloaded/printed by Delaware Univ (Delaware Univ) pursuant to License Agreement. No further reproductions authorized.

54

POLYMERIC COMPOSITES: SECOND VOLUME

Glass Transition Temperature Glass transition temperature T~ was measured by dynamic mechanical analysis (DMA). Samples were cut from conditioned and unconditioned unidirectional panels to roughly 6 by 1.25 cm using a diamond saw, and samples were taken at least 1 cm from the edge of a conditioned panel to avoid edge effects. Conditioned samples were tested after reaching moisture saturation levels in 20 and 80~ water baths and in the 20~ 75% RH environment, with the goal of producing bounds on the observed Tu reductions. DMA testing involved dynamic loading of samples in flexure (3-point bend) with a frequency of 1 Hz as the temperature was increased at a rate of 2~ T~ values were measured with fibers oriented transversely and longitudinally to the neutral axis of the specimen. T~, is reported as the peak in the loss modulus, E".

Damage Development Ultrasonic c-scans were made of cross-ply panels immersed in 20, 40, and 80~ water for at least 1300 h. The scans included normal incidence and 30 ~ polar backscattering, which enables determinations of transverse cracks [13]. After observing the onset of microcracking in the edge-polished samples under an optical microscope, moisture weight gain specimens were removed periodically from their conditioning environments, sectioned with a diamond saw, mounted in epoxy, ground and polished to a 5-txm finish. These polished cross-sectional surfaces of conditioned (and also unconditioned) samples were then observed using an optical microscope (magnifications up to 1000X), Damage was characterized as the number of transverse microcracks per ply per centimeter of cross-sectional material. Thus, a microcrack was counted only if it extended through the thickness of at least one ply.

Results

Characterization of Moisture Diffusion Absorption and desorption of moisture for the IM7/K3B composite system can be characterized, in general, by Fick's second law [2]. In Figs. 2 and 3, average moisture uptake data are shown for 20~ conditioning of unidirectional specimens and 40~ conditioning of thick cross-ply specimens, respectively. The fitted curves were generated using an approximate solution to Fick's second law [1]. Generally, the standard deviation for each data point and the scatter of the data points about the curves are within the experimental accuracy of the balance used to weigh the samples, for example, _+ 1 mg corresponds to __0.017 and ___0.015 wt%, respectively, for unidirectional and thick cross-ply specimens near saturation levels. However, a significant increase in the saturation levels of thick cross-ply specimens conditioned in 97 and 100% RH environments was observed after 4400 h (see Fig. 3). To date, moisture diffusion into unidirectional laminates has not deviated from Fickian behavior in any of the environments. (Conditioning in 40~ 75% RH and in 80~ 85 and 95% RH has not yet been completed.) Deviations from Fickian behavior, consisting of increases in saturation level after long times, have been observed during elevated-temperature conditioning of some cross-ply and quasi-isotropic specimens (see Discussion section). Moisture saturation levels M~, for the four laminate types increased with increasing humidity level, as shown in Fig. 4. This dependence was fitted with a power law function of the form [1,2] M~., = c(% RH)" Copyright by ASTM Int'l (all rights reserved); Mon Feb 22 15:54:11 EST 2016 Downloaded/printed by Delaware Univ (Delaware Univ) pursuant to License Agreement. No further reproductions authorized.

(l)

VANLANDINGHAM ET AL. ON THERMOPLASTIC POLYIMIDE COMPOSITES

55

0.6 0,5

o~

~_. E : - - - - ~ ---~

--

0.4 o.3 -

, ~ ."

~,~C - ~ ~'~" -d "~ r/"El' ~ . . ~" - ~

m-

" "t5 . . . . . . . .

[]

b"

o

v

,,Ill

0.20 1 9 - / ~

~)4~.~5 f ~ ' ' .,.~' 6,6 "~'~>'~-

o 75%RH --t:,- 85*/. RH --'- 9 7 0 RH --~ 100% RH

o,_J"

I

i

I

I

I

I

0

10

20

30

40

50

60

I I |

I

70

Time1/2 (hl/2) FIG. 2--Moisture absorption curves for unidirectional IM7/K3B specimens conditioned in 20~ environments. Each data point is an average o f 4 to 7 specimens. The curves shown approximate Fickian Behavior

~ 0.5

0.2 o~

I-e" 89%R H I

~-~

0

0

I " '~176176176 .. I I

I

I

I

I

I

10

20

30

40

50

60

70

Time1/2 (hl/2) FIG. 3--Moisture absorption curves for thick cross-ply IM7/K3B specimens conditioned in 40~ environments. Each data point is an average o f 9 specimens. The curves shown approximate Fickian behavior. Copyright by ASTM Int'l (all rights reserved); Mon Feb 22 15:54:11 EST 2016 Downloaded/printed by Delaware Univ (Delaware Univ) pursuant to License Agreement. No further reproductions authorized.

56

POLYMERIC COMPOSITES: SECOND VOLUME

~176 E,o],o 0.4 L

o•.

~

[45/0/'45/90]s [02/902/02]s

o

0.3 t~

=E

0.2 0.1 ~ M 0

0

s I

20

~ i

I

40

60

I

80

1O0

% RH FIG. 4--Moisture saturation levels plotted as a function of relative humidity for IM7/K3B laminates. Each data point is an average o f 3 to 9 specimens.

where c = 0.0011 and n = 1.34 are empirical parameters. No dependence of the saturation level on temperature has been observed over the investigated temperature range. Diffusivities were calculated for absorption in 20, 40, and 80~ environments and for desorption at 120~ using the solution for Fick's second law at short times [1,2]. A plot of diffusivity D as a function of inverse temperature is shown in Fig. 5. The diffusivity can be characterized by classical Arrhenius behavior and is given as a function of temperature, T (K), by D = 0.01 exp [ - 4 7 5 0 / T ]

(2)

Deviations from this diffusion behavior occurred during the desorption and re-absorption of moisture in cross-ply specimens. Desorption diffusivities of cross-ply specimens conditioned for 2400 h in 20~ environments and for 1200 to 1800 h in 40~ environments were 20 to 50% higher than those of unidirectional specimens that had received the same conditioning. Upon reabsorption of cross-ply specimens conditioned at 40~ absorption diffusivities increased 30 to 65% above their initial values. Thin cross-ply specimens conditioned for 2400 h in 20~ environments showed similar increases in absorption diffusivities, but the diffusivities of thick cross-ply specimens that had received the same conditioning remained relatively constant. Moisture cycling seems to also have a moderate effect on the diffusion behavior of unidirectional laminates conditioned for 1800 at 40~ as diffusivities increased between 15 and 25% above initial values. In several cases, slight increases (5 to 10%) in moisture saturation levels were also observed for both unidirectional and cross-ply specimens during re-absorption. A comparison of diffusivities and saturation levels between the first and second absorption cycles is shown in Table 1. Copyright by ASTM Int'l (all rights reserved); Mon Feb 22 15:54:11 EST 2016 Downloaded/printed by Delaware Univ (Delaware Univ) pursuant to License Agreement. No further reproductions authorized.

VANLANDINGHAM ET AL. ON THERMOPLASTIC POLYIMIDE COMPOSITES

10 "7

57

A

_

exp(-4750/T)

10.8

= [02/90]s t't

10-9

I

~

9

9 [45/0/-45/90]s

A [021902/02] s 10.10 0.0025

I I 0.0030

0.0035

1/T (K "1) FIG. 5--Arrhenius plot of diffusivity D (cme/s) as a function of inverse temperature for IM7/K3B laminates. Each data point is an average of 3 to 9 specimens.

Mode I Intralaminar Fracture Toughness

The results of fracture toughness testing of samples conditioned at 20 and 40~ are shown in Figs. 6 and 7, respectively. Unidirectional specimens conditioned in room-temperature, high-humidity environments for up to 3600 h have exhibited no measurable toughness loss. Similarly, fracture toughness values have remained constant for specimens conditioned in 40~ environments for up to 3000 h. Also, no toughness loss has been observed for specimens conditioned in 80~ 75% RH for up to 1500 h. Specimens conditioned to saturation (300 h) in 80~ 100% RH did exhibit a moderate loss in toughness of 30 to 40%. However, the degradation mechanism (unknown at this time) responsible for this toughness loss might be active only during elevated-temperature water immersion and may not be representative of an "accelerated" conditioning effect. Glass Transition Temperature

To date, only limited glass transition temperature Tg testing has been performed, the results of which are shown in Table 2. Standard deviations of the given values are of the order of 0.5~ Tg values for dry specimens are similar to a Tg value of 255~ reported by Cornelia [4] for the same composite system. T~ values for moisture-saturated specimens were as much as 11% lower than the baseline value for a sample conditioned in the 80~ 100% RH environment. However, this effect was reversible, as T~ returned to its baseline value when the specimen was redried. These results indicate that moisture acts as a plasticizer to the polymer matrix. Similar results have been obtained by Cornelia [4] for several polyimide composite systems. More extensive characterization of Tg as a function of the conditioning time, temperature, and humidity level is currently under evaluation (see Discussion section). Copyright by ASTM Int'l (all rights reserved); Mon Feb 22 15:54:11 EST 2016 Downloaded/printed by Delaware Univ (Delaware Univ) pursuant to License Agreement. No further reproductions authorized.

58

POLYMERIC COMPOSITES: SECOND VOLUME

TABLE l--Comparison of diffusivities D and moisture saturation levels M~,,for IM7/K3B specimens

selected for limited cycling studies.

Msat,

M~at,

Laminate/Conditioning Environment

Conditioning Time, h

D, cm2/s, Cycle 1

wt %, Cycle 1

D, cm2/s, Cycle 2

wt %, Cycle 2

[90ho 89% RH [90ho 40~ 89% RH [90], 0 40~ 97% RH [90]., 40~ 97% RH [90],0 40~ 100% RH [02/90]s 40~ 89% RH [02/90]s 40~ 97% RH [02/90]s 40~ 100% RH [02/902/02]s 40~ 89% RH [02/902/02]s 40~ 89% RH [0J902/0z]s 40~ 97% RH [02/90J0z]s 40~ 97% RH [02/902/02]s 40~ 100% RH [90ho 20~ 75% RH [90ho 20~ 85% RH [90],~ 20~ 97% RH [0z/90]s 20~ 75% RH [02/90]s 20~ 85% RH [02/90]s 20~ 97% RH [02/902/02]s 20~ 75% RH

1200

2.05 • 10 9

0.430

2.36 x 10 9

0.460

1800

1.81 X 10 -9

0.445

2.25 X 10 -9

0.465

1200

2.90 • 10 -9

0.470

2.80 x 10 9

0.490

1800

2.50 •

10 9

0.465

3.20 • 10 9

0.510

1800

2.40 X 10 9

0.510

2.80 X 10 9

0.540

1800

2.80 X 10 -9

0.470

2.82 • 10 -9

0.470

1800

2.60 X 10 ~

0.500

3.50 x 10 -~

0.540

1800

3.20 X 10 9

0.510

4.06 x 10 9

0.540

1200

1.96 X 10 9

0.450

2.60 X 10 9

0.450

1800

2.80 X 10 9

0.470

2.82 X 10 -9

0.470

1200

2.59 x 10 9

0.480

3.89 • 10 9

0.530

1800

2.60 • 10 9

0.500

3.50 • 10 9

0.540

1800

3.20 x 10 9

0.510

4.06 x 10 -9

0.540

2400

0.85 x 10 9

0.345

0.86 • 10 -9

...

2400

0.94 X 10 -9

0.420

1.15 X 10 -9

...

2400

0.86 • 10 -9

0.495

0.91 • 10 -9

...

2400

1.31 • 10 9

0.380

1.63 • 10 9

0.410

2400

0.83 • 10 -9

0.425

1.25 x l0 -9

0.440

2400

0.97 x 10 9

0.470

1.08 • 10 9

0.500

2400

1.17 X 10 -9

0.400

1.15 • 10 -9

...

[02/902/02]s

2400

1.24 x 10 -9

0.470

1.18 • 10 -9

...

20~ 85% RH [0z/902/02]s 20~ 97% RH

2400

0.95 x 10 9

0.520

1.02 • 10 9

...

40~

Damage Development Preliminary moisture R H for 1300 to 1500 F i c k i a n b e h a v i o r u p to level, as r e p o r t e d in a

studies i n c l u d e d c o n d i t i o n i n g o f c r o s s - p l y l a m i n a t e s in 80~ 100% h. D u r i n g c o n d i t i o n i n g , m o i s t u r e a b s o r p t i o n w a s c h a r a c t e r i z e d by s a t u r a t i o n f o l l o w e d b y a s e c o n d stage u p t a k e to a n e w s a t u r a t i o n p r e v i o u s section. A f t e r c o n d i t i o n i n g , b o t h thin and thick c r o s s - p l y

Copyright by ASTM Int'l (all rights reserved); Mon Feb 22 15:54:11 EST 2016 Downloaded/printed by Delaware Univ (Delaware Univ) pursuant to License Agreement. No further reproductions authorized.

59

VANLANDINGHAM ET AL. ON THERMOPLASTIC POLYIMIDE COMPOSITES

1400 1200 1000

A

800 O

I

i

04

600

o ~ , x

400 200 0

0

!

9 Baseline (dry) |

m

20~ 20~ 20~ 20oC

75% RH 85% RH 97% RH 100% RH

I

I

I

1000

2000

3000

4000

Conditioning Time (h) FIG. 6--1ntralaminar fracture toughness Gic (J/m 2) o f unidirectional IM7/K3B specimens conditioned in 20~ environments for up to 3600 h. Each data point is an average o f 4 specimens. No toughness loss was observed.

1400 1200 1000

A

04

-.j O

800 600

9 Baseline(dry) o 40~ = 40~ 9 7 % R H

(3 m

400

9 40~

200 0

0

100%RH

I

I

I

1000

2000

3000

4000

Conditioning Time (h) FIG. 7--1ntralaminar fracture toughness GIc (J/m 2) o f unidirectional IM7/K3B specimens conditioned in 4(TC environments. Each data point is an average of 4 specimens. No toughness loss was observed. Copyright by ASTM Int'l (all rights reserved); Mon Feb 22 15:54:11 EST 2016 Downloaded/printed by Delaware Univ (Delaware Univ) pursuant to License Agreement. No further reproductions authorized.

60

POLYMERIC COMPOSITES: SECOND VOLUME

TABLE 2--Tg measured after conditioning unidirectional IM7/K3B samples in various environments to saturation, and then again after redrying the samples. Conditioning Baseline, dry 20~ 75% RH ---* Redried 20~ 100% RH --, Redfied 80~ 100% RH ---. Redried

Tg Longitudinal Specimen, ~ 262~ 245 ~ 260 ~ 236 ~ 260~ 233~ 262 ~

Ts, Transverse Specimen, ~ 252~ 247~ 251 ~ 246~ 252~ 245 ~ 252~

laminates were sectioned, mounted in epoxy, and polished to reveal microcrack densities of 6 cracks/ply/cm or more. Subsequent conditioning studies have included periodic checks for microcracks. Except for a few anomalous specimens (see Discussion section), no significant microcracking (